SiamRPN++: Evolution of Siamese Visual Tracking with Very Deep Networks

TL;DR

This paper enhances Siamese visual trackers by addressing translation invariance issues, enabling the use of very deep networks like ResNet-50, leading to significant accuracy improvements and state-of-the-art results on multiple benchmarks.

Contribution

It introduces a spatial aware sampling strategy and a new architecture for depth-wise and layer-wise aggregation, advancing Siamese tracking performance with deep networks.

Findings

Achieved state-of-the-art results on four large benchmarks.

Proved translation invariance is key to deep network integration.

Reduced model size while improving accuracy.

Abstract

Siamese network based trackers formulate tracking as convolutional feature cross-correlation between target template and searching region. However, Siamese trackers still have accuracy gap compared with state-of-the-art algorithms and they cannot take advantage of feature from deep networks, such as ResNet-50 or deeper. In this work we prove the core reason comes from the lack of strict translation invariance. By comprehensive theoretical analysis and experimental validations, we break this restriction through a simple yet effective spatial aware sampling strategy and successfully train a ResNet-driven Siamese tracker with significant performance gain. Moreover, we propose a new model architecture to perform depth-wise and layer-wise aggregations, which not only further improves the accuracy but also reduces the model size. We conduct extensive ablation studies to demonstrate the effectiveness of the proposed tracker, which obtains currently the best results on four large tracking benchmarks, including OTB2015, VOT2018, UAV123, and LaSOT. Our model will be released to facilitate further studies based on this problem.